8 TUi: IXTEIiXAI. WoKK OF THE WIND. 



The wind velocities during this period of observation ranged from about 10 to 

 25 miles an hour, and the frequency of measurement was every 7 to 17 seconds. 

 If. mi the one hand, owing to the weigh! and inertia <>[' the anemometer, this is far 

 from dome- justice to the actual irregularities of the wind; on the other, it equally 

 shows that the wind was far from being a body of even approximate uniformity of 

 motion, and that, even when considered in quite small sections, the motion was 

 found to be irregular almost beyond conception, — certainly beyond anticipation; 

 for this record is not selected to represent an extraordinary breeze, but the normal 

 movement of an ordinary one. 



By an application of these facts, to be presented later, I then reached by these 

 experiments the conclusion that it was theoretically possible to cause a heavy body 

 wholly immersed in the wind to be driven in the opposite direction, e. g., to move 

 east while the wind was blowing west, without the use of any power other than 

 that which the wind itself furnished, and this even by the use of plane sur- 

 faces, and without taking advantage of the more advantageous properties of curved 

 ones. 



This power, I further already believed myself warranted by these experiments 

 in saying, could be obtained by the movements of the air in the horizontal plane 

 alone, even without the utilization of currents having an upward trend. But I was 

 obliged to turn to other occupations, and did not resume these interesting observa- 

 tions until the year 1893. 



Although the anemometer used at Allegheny served to illustrate the essential 

 fact of the rapid and continuous fluctuations of even the ordinary and comparatively 

 uniform wind, yei owing to the inertia of the arms and cups, which tended to 

 equalize the rate (the moment of inertia was approximately 40,000 gr. cm. 2 ), and to 

 the fact that the record was only made at every twenty-fifth revolution, the internal 

 changes in the horizontal component of the wind's motion, thus representing its 

 potential work, were not adequately recorded. 



In January, 189.'!, 1 resumed these observations at Washington with apparatus 

 with which I sought to remedy these defects, using as a station the roof of the 

 north Toucr of the Smithsonian Institution building, the top of the parapet being 

 14'J feet i }:;.."> metro) above the ground, and the anemometers, which were located 

 above the parapet, being L53 feet (46.7 metres) above the ground. I placed them in 

 charge of Mr. George E. Curtis, with instructions to take observations under the 

 conditions of light, moderate, and high winds. The apparatus used w as, first, a 

 Weather Bureau Robinson anemometer of standard size, with aluminum cups. 

 I >iameter to centre of cups :!4 cm. ; diameter of cups 10.16 cm. ; weight of arms and 

 cups 241 grammes; approximate moment of inertia, 40,710 gr. cm." 



